The impact of treatment and treatment interactions on employment outcomes for individuals with schizophrenia and other severe mental disorders.

摘要

Hypothesis. The aim of this research is to investigate interactions between vocational rehabilitation and other treatments for patients with schizophrenia in promoting employment outcomes.; Methods. Medical evaluation and maintenance (MEM), psychotherapy (PT), and case management (CM) were the key non-vocational treatment variables selected for this analysis. Regression analysis was used to explore the impact of these treatments and the interaction of these treatments with IPS on employment. In addition, instrumental variables (IV) two-staged least squares regression (2SLS) and bivariate probit models were used to address the potential for endogenous treatment.; Results. The results of the first set of IV models, treating all the treatment variables as endogenous, indicate that PT has a negative impact on employment, CM has a mixed impact on employment, and the interaction of IPS and PT has a negative impact on employment. However, the results indicate that IPS, the interaction of IPS and CM, and the interaction of IPS and MEM can have a positive impact on employment. The results of the second set of IV models, which use the assumption that PT is the only endogenous variable, still indicate that PT has a negative impact on employment, as does the interaction of IPS and PT. In addition, the results of the bivariate probit models that treat PT as endogenous still indicate that PT has a negative impact on employment, and IPS, CM, MEM, the interaction of IPS and PT, and the interaction of IPS and MEM can have a positive impact on employment.; Conclusion. The most potentially useful results from this analysis are from the PT bivariate probit models, which indicate that PT has a negative impact on employment and IPS, CM, and MEM and the interactions of IPS with PT and MEM can each have a positive impact on employment. The reason that the coefficients for PT are consistently negative is not obvious, and negative coefficients for PT seem counter-intuitive. Potential explanations include that the time commitment of repeatedly going to PT treatment sessions interferes with employment, omitted variable bias, and that PT may be detrimental to employment.